Human 60 kDa heat shock protein is a target autoantigen of T cells derived from atherosclerotic plaques.

of February 6, 2017.

This information is current as

Atherosclerotic Plaques

Target Autoantigen of T Cells Derived from

Gianfranco Del Prete

Gianni Rombolà, Sergio Romagnani, Antonio Cassone and

Annalisa Azzurri, Ruurd van der Zee, Alessandra Ciervo,

Marisa Benagiano, Mario M. D'Elios, Amedeo Amedei,

http://www.jimmunol.org/content/174/10/6509

doi: 10.4049/jimmunol.174.10.6509

2005; 174:6509-6517; ;

J Immunol

References

http://www.jimmunol.org/content/174/10/6509.full#ref-list-1

, 24 of which you can access for free at:

cites 43 articles

This article

Subscriptions

http://jimmunol.org/subscriptions

is online at:

The Journal of Immunology

Information about subscribing to

Permissions

http://www.aai.org/ji/copyright.html

Submit copyright permission requests at:

Email Alerts

http://jimmunol.org/cgi/alerts/etoc

Receive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606.

Immunologists All rights reserved.

Copyright © 2005 by The American Association of

9650 Rockville Pike, Bethesda, MD 20814-3994.

The American Association of Immunologists, Inc.,

is published twice each month by

The Journal of Immunology

by guest on February 6, 2017 http://www.jimmunol.org/ Downloaded from by guest on February 6, 2017 http://www.jimmunol.org/ Downloaded from

Human 60-kDa Heat Shock Protein Is a Target Autoantigen of

T Cells Derived from Atherosclerotic Plaques

1

Marisa Benagiano,

2

* Mario M. D

’

Elios,

2

*

†

Amedeo Amedei,*

†

Annalisa Azzurri,*

Ruurd van der Zee,

‡

_{Alessandra Ciervo,}

§

_{Gianni Rombola`,}

†

_{Sergio Romagnani,*}

†

Antonio Cassone,

§

and Gianfranco Del Prete

3

*

†

Epidemiological studies suggest the potential importance of an inflammatory component in atherosclerosis and support the hypothesis that immune responses to Ags of pathogens cross-react with homologous host proteins due to molecular mimicry. Protein candidates involved may be the stress-induced proteins known as heat shock proteins (HSP). In this study, we report that atherosclerotic plaques harbor in vivo-activated CD4ⴙT cells that recognize the human 60-kDa HSP. Such in vivo-activated 60-kDa HSP-specific T cells are not detectable in the peripheral blood. In patients with positive serology and PCR for Chlamydia

pneumoniae DNA, but not in patients negative for both, most of plaque-derived T cells specific for human 60-kDa HSP also

recognized the C. pneumoniae 60-kDa HSP. We characterized the submolecular specificity of such 60-kDa HSP-specific plaque-derived T cells and identified both the self- and cross-reactive epitopes of that autoantigen. On challenge with human 60-kDa HSP, most of the plaque-derived T cells expressed Th type 1 functions, including cytotoxicity and help for monocyte tissue factor production. We suggest that arterial endothelial cells, undergoing classical atherosclerosis risk factors and conditioned by Th type 1 cytokines, express self 60-kDa HSP, which becomes target for both autoreactive T cells and cross-reactive T cells to microbial 60-kDa HSP via a mechanism of molecular mimicry. This hypothesis is in agreement with the notion that immunization with HSP exacerbates atherosclerosis, whereas immunosuppression and T cell depletion prevent the formation of arteriosclerotic lesions in experimental animals. The Journal of Immunology, 2005, 174: 6509 – 6517.

A

therosclerosis is a multifactorial disease for which a number of different pathogenic mechanisms have been proposed. In the last two decades, attention has been given to the inflammatory processes associated with atherogenesis. In addition to classical risk factors for atherosclerosis, such as high levels of low-density lipoprotein (LDL)4 _{cholesterol or oxidized}

LDL, free radicals, hyperglycemia, and genetic susceptibility to endothelial damage (1, 2), a pathogenic role for infections in ath-erosclerosis is suggested by the detection of pathogens (e.g.,

Chla-mydia pneumoniae, CMV, or herpes viruses) in the arterial vessels

and the association between atherosclerosis and increased Ab lev-els to these pathogens (3). Observations in humans and animals

suggest that atherosclerotic plaques derive from specific cellular and molecular mechanisms that can be ascribed to an inflammatory disease of the arterial wall, the lesions of which consist of mac-rophages and T lymphocytes (4 – 6). Activated macmac-rophages and T cells would be responsible for in situ production of enzymes, cy-tokines, and chemokines that further expand the process. If inflam-mation continues unabated, it results in increased numbers of plaque-infiltrating macrophages and T cells, which contribute to remodeling of the arterial wall (7). Within the T cell population of the plaque, most of the cells are activated CD4⫹cells expressing HLA-DR and CD25 of the IL-2R (8). C. pneumoniae DNA and C.

pneumoniae-specific T cell clones were found in the plaques of

anti-C. pneumoniae seropositive patients. The specificity reper-toire of such CD4⫹T cells included the C. pneumoniae 60-kDa heat shock protein (CpHSP60), the 10-kDa HSP, the outer mem-brane protein 2, or undefined Ags of the C. pneumoniae elemen-tary bodies. T cell clones recovered from C. pneumoniae DNA-negative plaques of anti-C. pneumoniae seroDNA-negative patients did not react to C. pneumoniae Ags (9).

In this study, we focused on the analysis of the T cell infiltrates of atherosclerotic plaques of anti-C. pneumoniae seronegative pa-tients, making the hypothesis that human proteins expressed under stress conditions, such as the human 70-kDa HSP (hHSP70) or human 60-kDa HSP (hHSP60), might be autoantigens recognized by plaque-infiltrating T cells.

Materials and Methods

Patients

Carotid plaques were obtained by endoarterectomy from eight patients (five males and three females; mean age, 67 years; range, 62–73 years) with atherosclerotic arteriopathy. Four patients were seronegative for

anti-C. pneumoniae Abs (Cp-neg), whereas the other four patients were

sero-positive for anti-C. pneumoniae Abs (Cp-pos), as shown by both commer-cial ELISA tests (Eurospital) and standard microimmunofluorescence

*Department of Internal Medicine, University of Florence, and†

Laboratory of Im-munogenetics, Department of Biomedicine, Azienda Ospedaliero-Universitaria di Careggi, Florence, Italy;‡

Department of Infectious Diseases and Immunology, Fac-ulty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands; and§

De-partment of Infectious, Parasitic and Immune-Mediated Diseases, Istituto Superiore di Sanita`, Rome, Italy

Received for publication September 17, 2004. Accepted for publication February 16, 2005.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by grants from the Italian Ministry of University and Research, the Ministery of Health, the University of Florence, the Associazione Itali-ana per la Ricerca sul Cancro, and the Istituto Superiore di Sanita`.

2

M.B. and M.M.D. contributed equally to this publication.

3

Address correspondence and reprint requests to Dr. Gianfranco Del Prete, Depart-ment of Internal Medicine, Viale Morgagni 85-50134 Florence, Italy. E-mail address: gdelprete@unifi.it

4

Abbreviations used in this paper: LDL, low-density lipoprotein; HSP, heat shock protein; CpHSP60, C. pneumoniae 60-kDa HSP; hHSP70, human 70-kDa HSP; hHSP60, human 60-kDa HSP; neg, seronegative for anti-C. pneumoniae Ab; Cp-pos, seropositive for anti-C. pneumoniae Ab; MI, mitogenic index; BCG, bacillus Calmette-Gue´rin; MbHSP65, Mycobacterium bovis BCG 65-kDa HSP; EBV-B, EBV transformed B; TF, tissue factor.

Copyright © 2005 by The American Association of Immunologists, Inc. 0022-1767/05/$02.00

by guest on February 6, 2017

http://www.jimmunol.org/

assay (cutoff value, 32). Samples of PBMC were obtained from each pa-tient. Their MHC haplotypes were as follows: HLA-A2, B7, B51, DRB1*14, and DRB1*16 in patient 1; HLA-A1, A2, B13, B18, DRB1*03, and DRB1*07 in patient 2; HLA-A24, B15, B40, DRB1*01, and DRB1*16 in patient 3; HLA-A1, A2, B18, DRB1*04, and DRB1*11 in patient 4; HLA-A24, B44.03, B58.01, DRB1*0701, and DRB1*16 in patient 5; HLA-A2, B15, B44, and DRB1*13 in patient 6; HLA-A1, B8, B27, DRB1*03, and DRB1*07 in patient 7; and HLA-A2, A24, B7, B51, DRB1*0701, and DRB1*11 in patient 8.

Detection of C. pneumoniae in atherosclerotic plaques

The presence of C. pneumoniae was investigated by nested PCR, as reported elsewhere (9, 10). Briefly, DNA was extracted from fragments of all the en-doarterectomy specimens by QIAamp DNA kit (Qiagen). Nested PCR con-sisted of two rounds of amplification using two sets of primers, each in a 50-␮l volume. On completion of primary PCR (37 cycles), 2␮l of the PCR product were added into fresh reaction mix containing the second set of primers and amplified for 25 cycles. The amplified DNA products were analyzed by elec-trophoresis in 1.5% agarose gel, stained with ethidium bromide, and hybrid-ized as reported previously (10). The nested PCR for C. pneumoniae included an outer primer pair (HL-1, HR-1) and an inner pair (HM-1, HR-2) that gen-erated a product of 204 bp. The details of primers and probe are as follows: HL-1, -5⬘-GTTGTTCATGAAGGCCTACT-3⬘ end; HR-1, -5⬘-TGCATAAC CTACGGTGTGTT-3⬘ end; HM-1, -5⬘-GTGTCATTCGCCAAGGTTAA-3⬘ end; HR-2, -5⬘-ACCTGTCCAAGGTTCATCCT-3⬘ end; and DNA probe, -5⬘-GTGTCATTCGCCAAGGTTAAAGTCTACGTT-3⬘ end.

Generation of T cell clones from atherosclerotic plaques and peripheral blood

Fragments of atherosclerotic plaques and samples of PBMC were cultured for 7 days in RPMI 1640 medium supplemented with IL-2 to expand in vivo-activated T cells. Single T cell blasts were then cloned under limiting dilution (9, 11–13). Briefly, single T cell blasts were seeded in microwells (0.3 cells/well) in the presence of 2⫻ 105_{irradiated (5000 rad) allogeneic}

PBMC, PHA (0.5% vol/vol), and IL-2 (50 U/ml). At weekly intervals, irradiated allogeneic PBMC and IL-2 were added to each microculture to maintain the expansion of growing clones. Ag specificity of T cell clones was assessed by measuring [3_{H]thymidine uptake after 60 h of coculture}

with irradiated autologous PBMC in the presence of medium, recombinant hHSP70 (Sigma-Aldrich) (10 ␮g/ml), recombinant hHSP60 (Sigma-Al-drich) (10␮g/ml), or recombinant CpHSP60 (10 ␮g/ml), prepared as en-dotoxin-free material (14). No significant proliferation of T cell clones was found in response to irradiated autologous PBMC alone. The mitogenic index (MI) was calculated as the ratio between mean values of cpm ob-tained in stimulated cultures and those obob-tained in the presence of medium alone. MI⬎5 was considered positive.

T cell clones reactive to hHSP60 were also tested for proliferation in response to the recombinant HSP65 protein of Mycobacterium bovis ba-cillus Calmette-Gue´rin (BCG) (MbHSP65) (Aalto Bio Reagents) and to the recombinant 60-kDa chaperonin GroEL of Escherichia coli (Sigma-Al-drich) (10␮g/ml).

All of the 26 plaque-derived T cell clones reactive to hHSP60 and the 18 clones reactive to both hHSP60 and CpHSP60 expressed a CD3⫹CD4⫹CD8⫺ phenotype and showed a single peak of fluorescence intensity. The repertoire of the TCR V␤ chain of HSP60-specific T cell clones was analyzed with a panel of 22 mAbs specific to the following: V␤1, V␤2, V␤4, V␤7, V␤9, V␤11, V␤14, V␤16, V␤18, V␤20, V␤21.3, V␤22, and V␤23 (Beckman Coulter Immunotech); and V␤3.1, V␤5.1, V␤5.2, V␤5.3, V␤6.7, V␤8, V␤12, V␤13, and V␤17 (AMS Biotechnol-ogy); and isotype-matched nonspecific Ig were used as negative control. Data acquisition was performed in a FACSCalibur flow cytometer using the CellQuest software program (BD Biosciences). From each T cell clone, mRNA was extracted by mRNA direct isolation kit (Qiagen). For cDNA synthesis, the same amount of mRNA (50 ng) was used, and cDNA was synthesized by Moloney murine leukemia virus-reverse transcriptase (New England Biolabs) and oligo(dT) primers according to enzyme supplier’s protocol. cDNA mix of all samples was amplified under equal conditions by a 30-cycle PCR using a V␤ TCR typing amplimer kit for V␤10, V␤15, and V␤19 (BD Clontech) according to the manufacturer’s instructions.

Submolecular specificity of plaque-derived T cell clones reactive to hHSP60 or to both hHSP60 and CpHSP60

To span the 573 aa sequence of hHSP60 and the 544 aa sequence of CpHSP60, 113 and 107 overlapping 15-mer peptides with a 10 (5 on each side) aa overlap, respectively, were prepared by automated, simultaneous multiple peptide synthesis, as described previously (15) Homologies

be-tween the two series of peptides were screened by using the basic local alignment search tool server of the National Center for Biotechnology Information.

Two series of 15-mer peptides corresponding to a number of sequences of the HSP65 protein of M. bovis BCG (MbHSP65) and of the 60-kDa chaperonin GroEL of E. coli (aa 21–35, 26 – 40, 31– 45, 46 – 60, 51– 65, 56 –70, 121–136, 126 –140, 131–145, 141–155, 146 –160, 151–165, 161– 175, 166 –180, 171–185, 181–195, 186 –200, 191–205, 211–225, 216 –230, 221–235, 406 – 420, 411– 425, 416 – 430, 421– 435, and 426 – 440) were also prepared. Equal amounts of each component of the two series of over-lapping peptides of hHSP60 and of CpHSP60 were pooled to have two series of 11 peptide pools. T cell blasts (4⫻ 104_{) from each clone were}

cultured in triplicate for 3 days together with irradiated autologous mono-nuclear cells (1.5⫻ 105_{) in the presence of medium, hHSP60 (10␮g/ml),}

CpHSP60 (10␮g/ml), or equal aliquots from each of the 22 pools in which each peptide component was present at a 10 ␮g/ml final concentration. After 60 h, [3_{H]TdR uptake was measured. T cell blasts of each clone were}

then retested for proliferation to the individual peptide components of the pool that had induced a MI⬎5.

MHC class II restriction of hHSP60 epitope recognition by plaque-derived T cell clones

The effect of anti-HLA-DR (clone G46 – 6) or anti-HLA-DQ (clone TU169; BD Biosciences Pharmingen) (5␮g/ml final concentration) mAbs or their isotype control (mouse IgG2a) on T cell clone proliferation induced by hHSP60 or CpHSP60 was assessed. The MHC class II restriction of the proliferative response of T cell clones to hHSP60 or CpHSP60 peptides was assessed by using irradiated allogeneic APC. To this end, PBMC from both patients and healthy donors sharing with patients one of the DRB1* alleles were stimulated with PHA followed by IL-2 to obtain polyclonal lines of activated T cells to be used as irradiated (3000 rad) allogeneic APC in coculture experiments with T cell clones in the presence of the HSP60 peptide to which they reacted.

Assessment of the cytokine profile of T cell clones

To assess the cytokine production of hHSP60-specific clones on Ag stim-ulation, 5⫻ 105_{T cell blasts of each clone were cocultured for 48 h in 0.5}

ml of medium with 5⫻ 105_{irradiated autologous PBMC in the absence or}

presence of hHSP60 or CpHSP60 (10␮g/ml). At the end of culture period, duplicate samples of each supernatant were assayed for IFN-␥, TNF-␣, and IL-4 (BioSource International) (9, 13). T cell clones able to produce IFN-␥, but not IL-4, were categorized as Th1; clones able to produce IL-4, but not IFN-␥, were categorized as Th2; and clones producing both IFN-␥ and IL-4 were categorized as Th0.

Perforin-mediated cytotoxicity and Fas-Fas ligand-mediated proapoptotic activity

Perforin-mediated cytolytic activity of T cell clones was assessed as re-ported previously (16). T cell blasts of hHSP60-specific clones were in-cubated at ratios of 10:1, 5:1, and 2.5:1 with51_{Cr-labeled autologous EBV}

transformed B (EBV-B) cells preincubated with hHSP60 (10␮g/ml). After centrifugation, microplates were incubated for 8 h at 37°C, and 0.1 ml of supernatant was removed for measurement of 51_{Cr release, as reported}

previously (17). The ability of hHSP60-specific T cell clones to induce Fas-Fas ligand-mediated apoptosis was assessed using Fas⫹Jurkat cells as target. T cell blasts from each clone were cocultured with51_{Cr-labeled Jurkat cells at}

E:T ratios of 10:1, 5:1, and 2.5:1 for 18 h in the presence of PMA (10 ng/ml) and ionomycin (1 mmol/l), as reported previously (17, 18).

Assay for T cell clone helper function for monocyte tissue factor (TF) production

T cell blasts of hHSP60-specific clones (8⫻ 105_{/ml) were cocultured for}

16 h with autologous monocytes (4⫻ 105_{/ml) in the presence of medium,}

hHSP70 (10␮g/ml), hHSP60 (10 ␮g/ml), or CpHSP60 (10 ␮g/ml). At the end of culture period, TF protein was measured by a specific ELISA (American Diagnostica) in duplicate samples of the supernatants obtained from cell suspensions after solubilization of membrane proteins with Triton X-100 and ultracentrifugation, as reported previously (19).

Results

Autoreactive CD4⫹T cell clones specific for hHSP60 are present in atherosclerotic lesions

In vivo-activated T cells resident in the plaques or in the peripheral blood were expanded in vitro in IL-2-conditioned medium and

by guest on February 6, 2017

http://www.jimmunol.org/

then cloned by a procedure that has proved useful and accurate for studies of tissue-infiltrating T cells in various diseases (9, 11–13). A total number of 151 CD4⫹and 23 CD8⫹T cell clones were obtained from the plaques of the four Cp-neg patients, whereas 115 CD4⫹ and 21 CD8⫹ were the T cell clones derived from the plaques of the four Cp-pos patients. Nested PCR on

endarterec-tomy specimens showed C. pneumoniae genomic material in each of the plaques obtained from the four Cp-pos patients but not in the plaques from the Cp-neg patients (data not shown).

For each patient, randomly selected CD4⫹and CD8⫹T cell clones derived from PBMC were matched to the corresponding plaque-de-rived T cell clones and assayed for proliferation in response to Table 1. Atherosclerotic plaques, but not peripheral blood, harbor CD4⫹T cells specific for hHSP60a

Patients and Source of T Cells

Total Number of CD4⫹T cell Clones Tested

Number (%) of Clones Reactive to hHSP60 Both hHSP60 and CpHSP60 C. pneumoniae-negative patients 1. Plaque 40 6 (15) 0 PBMC 40 0 0 2. Plaque 38 6 (16) 0 PBMC 38 0 0 3. Plaque 36 5 (14) 0 PBMC 36 0 0 4. Plaque 37 4 (11) 0 PBMC 37 0 0 C. pneumoniae-positive patients 5. Plaque 20 7 (35) 5/7 (71) PBMC 20 0 0 6. Plaque 25 3 (12) 3/3 (100) PBMC 25 0 0 7. Plaque 33 6 (18) 4/6 (67) PBMC 33 0 0 8. Plaque 37 7 (19) 6/7 (86) PBMC 37 0 0

a_{Equal numbers of clones derived from plaques or PBMC were screened for responsiveness to hHSP60 and CpHSP60 in} the presence of irradiated autologous mononuclear cells by measuring3_{[H]-thymidine uptake after 60 h.}

Table II. Epitope specificity of plaque-derived CD4 T cell clones reactive to hHSP60 but not to CpHSP60a

T Cell Clone (TCR V␤)

Proliferative Response (MI) to

Peptide Amino Acid Sequence (position) hHSP70 hHSP60 hHSP60 Peptide 1.10 (V␤11) ⬍2 36 49 MLRLPTVFRQMRPVS (1–15) 1.30 (V␤1) ⬍2 27 53 MRPVSRVLAPHLTRA (11–25) 1.54 (V␤17) ⬍2 21 23 VKDGKTLNDELEIIE (201–215) 1.26 (V␤5.1) ⬍2 92 118 LEIANAHRKPLVIIA (261–275) 1.43 (V␤18) ⬍2 102 97 TLNLEDVQPHDLGKV (331–345) 1.08 (V␤7) ⬍2 111 78 MAGDFVNMVEKGIID (506–520) 2.14 (V␤14) ⬍2 168 139 TVFRQMRPVSRVLAP (6–20) 2.08 (V␤20) ⬍2 69 84 KFGADARALMLQGVD (31–45) 2.58 (V␤21.3) ⬍2 46 52 KLVQDVANNTNEEAG (96–110) 2.40 (V␤5.2) ⬍2 91 68 NPVEIRRGVMLAVDA (136–150) 2.53 (V␤2) ⬍2 47 71 GGAVFGEEGLTLNLE (321–335) 2.11 (V␤6.7) ⬍2 63 88 MAGDFVNMVEKGIID (506–520) 3.18 (V␤8) ⬍2 473 439 TVFRQMRPVSRVLAP (6–20) 3.40 (V␤4) ⬍2 28 42 KNIGAKLVQDVANNT (91–105) 3.53 (V␤12) ⬍2 144 198 RRGVMLAVDAVIAEL (141–155) 3.43 (V␤3.1) ⬍2 19 28 TLNDELEIIEGMKFD (206–220) 3.24 (V␤16) ⬍2 467 488 MAGDFVNMVEKGIID (506–520) 4.13 (V␤9) ⬍2 126 184 MLRLPTVFRQMRPVS (1–15) 4.07 (V␤4) ⬍2 52 41 PYFINTSKGQKCEFQ (226–240) 4.31 (V␤22) ⬍2 29 33 EIIKRTLKIPAMTIA (466–480) 4.25 (V␤11) ⬍2 68 91 VEKIMQSSSEVGYDA (491–505) 5.02 (V␤8) ⬍2 55 63 MRPVSRVLAPHLTRA (11–25) 5.14 (V␤12) ⬍2 71 113 GGAVFGEEGLTLNLE (321–335) 7.31 (V␤14) ⬍2 80 98 TVFRQMRPVSRVLAP (6–20) 7.12 (V␤11) ⬍2 32 47 MAGDFVNMVEKGIID (506–520) 8.30 (V␤5.2) ⬍2 102 178 MLRLPTVFRQMRPVS (1–15) a

T cell blasts from each clone were cocultured with irradiated autologous APC in the presence of medium alone, hHSP70, hHSP60, or hHSP60 peptides (10␮g/ml), and proliferative responses (MI) were measured after 3 days. Results are reported as mean values obtained in quadruplicate cultures, SD values being⬍14% of means. Letters underlined in the recognized hHSP60 peptides indicate the amino acids shared with CpHSP60 along the sequence. Patients 1– 4 were anti-C. pneumoniae seronegative, whereas patients 5– 8 were seropositive.

by guest on February 6, 2017

http://www.jimmunol.org/

hHSP70, hHSP60, and CpHSP60. None of the CD8⫹clones derived from either plaques or PBMC showed proliferation to those Ags. Likewise, none of the 266 CD4⫹clones generated from the PBMC of either Cp-neg or Cp-pos patients showed significant proliferation to the Ags tested (Table I), although they proliferated in response to mitogen stimulation (data not shown). In contrast, a variable propor-tion between 11 and 35% of the CD4⫹T cell clones generated from plaque-infiltrating T cells of either Cp-neg or Cp-pos patients prolif-erated significantly to hHSP60 (Table I) but not to hHSP70 Ag (MI ⬍2). Under the same conditions, none of the 21 hHSP60-specific CD4⫹clones from the plaques of the four Cp-neg patients prolifer-ated significantly to CpHSP60. In contrast, in the series of the 23 hHSP60-specific CD4⫹clones from the plaques of the four Cp-pos patients, 18 (78%) proliferated equally well to both hHSP60 and CpHSP60 (Table I). Evidence for clonality of the CD3⫹CD4⫹CD8⫺ T cell clones, which is critical for interpretation of data, was provided by the cytofluorimetric patterns of single TCR-V␤ expression and the staining by only one of the TCR-V␤-chain-specific mAbs, with a single peak of fluorescence intensity (data not shown).

Submolecular specificity of autoreactive hHSP60-specific T cell clones

T cell blasts from each of the T cell clones reactive to hHSP60, but not to CpHSP60, were screened for proliferation in response to the 113 overlapping peptides for the hHSP60 (Table II). Each of these autoreactive T cell clones proliferated almost equally well to both the entire hHSP60 protein and to an epitope of such autoantigen. Interestingly, some hHSP60 epitopes, such as the 1–15, 6 –20, and 506 –520, were recognized by clones from different donors, despite their different MHC class II haplotypes or different TCR-V␤ ex-pression by T cell clones (Table II). Also, the five T cell clones recovered from atherosclerotic plaques of Cp-pos patients that pro-liferated to hHSP60, but not to CpHSP60, recognized the 1–15, 6 –20, 11–25, 321–335, and 506 –520 epitopes of hHSP60 (Table II). The lack of responsiveness to CpHSP60 of the 26 hHSP60-specific clones was confirmed by their inability to proliferate in response to the CpHSP60 peptide corresponding to the hHSP60 epitope to which they were reactive (data not shown), although a number of the hHSP60 epitopes recognized by this series of clones

shared a few (up to 7) amino acids with the corresponding CpHSP60 peptides (Table II).

Each of the 18 T cell clones that proliferated to both hHSP60 and CpHSP60 was screened for proliferation in response to the 113 overlapping peptides for the hHSP60 and the 107 peptides for the CpHSP60 (Table III). As expected, each of these cross-reactive T cell clones proliferated not only to both the entire hHSP60 protein and to an epitope of such autoantigen but also to the entire CpHSP60 and to a CpHSP60 peptide. However, the relevant cross-reactive CpHSP60 peptides showed high-sequence homology (from 7 to 12 of 15 aa) to the corresponding stimulatory hHSP60 epitopes. Interestingly, the panel of hHSP60 epitopes recognized by cross-reactive clones included 18 different peptides, with only some overlap, mainly in the 436 – 460 aa sequence for clones 7.03 and 7.22 of patient 7 and clone 8.14 of patient 8 (Table III).

Due to the high homology between the human and the bacterial 60-kDa HSPs, all of the hHSP60-specific clones were tested for their ability to proliferate in response to graded concentrations of the MbHSP65 of BCG or to the 60-kDa chaperonin GroEL of E.

coli. None of the 26 T cell clones reactive to hHSP60, but not to

CpHSP60, showed detectable proliferation to MbHSP65 or to GroEL, even at a dose as high as 50␮g/ml (data not shown). In contrast, 11 of the 18 T cell clones reactive to both hHSP60 and CpHSP60 also showed poor, but detectable, reactivity (range of MI, 2.8 –16.1) to comparable concentrations of MbHSP65. How-ever, at MbHSP65 concentrations⬍2␮g/ml, none of the 11 clones showed MI⬎2. Likewise, 9 of the same 18 clones showed poor reactivity to GroEL (range of MI, 2.4 –17.2), but at lower Ag doses (⬍2␮g/ml) no proliferation was detectable. Each of the 11 clones potentially cross-reactive to MbHSP65 and each of the 9 clones potentially cross-reactive to GroEL were also tested for prolifera-tion in response to the MbHSP65 and GroEL peptides that partially overlapped their specific hHSP60/CpHSP60 epitopes. As shown in Table IV, at a concentration of 1 ␮g/ml, neither MbHSP65 nor GroEL proteins or their potentially relevant peptides were able to induce T cell clone proliferation, whereas both hHSP60 and CpHSP60 and their relevant peptides were still stimulatory. It is interesting to note that a single amino acid difference (S instead of T at position 79 of hHSP60 or at position 55 of CpHSP60) accounted Table III. Epitope specificity of CD4 T cell clones reactive to both hHSP60 and CpHSP60 obtained from the atherosclerotic plaques of anti-C. pneumoniae seropositive patientsa

T Cell Clone (TCR V␤)

Proliferative Response (MI) to

hHSP60 hHSP60 Peptide (sequence) CpHSP60 CpHSP60 Peptide (sequence)

5.05 (V␤17) 38 81 (DGVTVAKSIDLKDKY 76–90) 73 89 (KDGVTVAKEIELEDK 51–65) 5.16 (V␤11) 75 113 (EEIAQVATISANGDK 166–180) 43 107 (HKEIAQVATISANND 141–155) 5.19 (V␤13.1) 49 37 (DAYVLLSEKKISSIQ 241–255) 67 44 (EDAL I LIYDKKISGI 216–230) 5.18 (V␤14) 20 33 (VGGTSDVEVNEKKDR 406–420) 31 39 (VGAATEIEMKEKKDR 381–395) 5.12 (V␤9) 34 61 (PTKVVRTALLDAAGV 521–535) 22 58 (LDPTKVTRSALESAA 496–510) 6.37 (V␤5.1) 107 76 (TVIIEQSWGSPKVTK 61–75) 88 51 (RHVVIDKSFGSPQVT 36–50) 6.08 (V␤14) 23 57 (VIAELKKQSKPVTTP 151–165) 21 48 (VVVDELKKISKPVQH 126–140) 6.41 (V␤8) 33 27 (LKVGLQVVAVKAPGF 291–305) 29 41 (RLRAGFRVCAVKAPG 266–280) 7.18 (V␤9) 44 61 (LLADAVAVTMGPKGR 46–60) 21 67 (KTLAEAVKVTLGPKG 21–35) 7.16 (V␤4) 89 113 (LEIIEGMKFDRGYIS 211–225) 147 131 (VLDVVEGMNFNRGYL 186–200) 7.03 (V␤1) 45 37 (GCALLRCIPALDSLT 441–455) 28 42 (GTALVRCIPTLEAFL 416–430) 7.22 (V␤13.2) 121 215 (RCIPALDSLTPANED 446–460) 120 91 (RCIPTLEAFLPMLAN 421–435) 8.22 (V␤18) 51 78 (VAVTMGPKGRTVIIE 51–65) 39 71 (AVKVTLGPKGRHVVI 26–40) 8.26 (V␤12) 136 104 (VATISANGDKEIGNI 171–185) 96 122 (QVATISANNDSEIGN 146–160) 8.01 (V␤5.2) 27 40 (KKVGRKGVITVKDGK 191–205) 23 31 (MEKVGKNGSITVEEA 166–180) 8.14 (V␤9) 76 68 (IVLGGGCALLRCIPA 436–450) 71 89 (ILPGGGTALVRCIPT 411–425) 8.07 (V␤22) 417 398 (VNMVEKGIIDPTKVV 511–525) 548 377 (AYTDMIDAGILDPTK 486–500) 8.18 0(V␤11) 142 201 (ASLLTTAEVVVTEIP 536–550) 170 226 (LLTTEALIADIPEEK 516–530) a

Culture conditions are reported in the legend to Table II. Underlined letters indicate the amino acids shared between hHSP60 and CpHSP60.

by guest on February 6, 2017

http://www.jimmunol.org/

for a strong reduction of stimulation of clone 5.05. Likewise, a single amino acid change (N instead of T at position 61 of hHSP60) hampered the recognition by clone 8.22 of the 26 – 40 epitope of either MbHSP65 or GroEL, whereas the substitution of T with H at position 37 of CpHSP60 did not affect cross-recognition.

The relative potency in inducing T cell clone proliferation of the self (hHSP60) and/or the corresponding cross-reactive CpHSP60 protein and peptides was assessed by comparison of dose-response curves. At 10␮g/ml, Ag proteins and peptides were almost equally potent in inducing T cell clone proliferation (Fig. 1 and Table IV). At lower doses (such as 1 or 0.1␮g/ml), the MI obtained with the appropriate self- or cross-reactive peptide was consistently higher than that obtained with the corresponding entire Ag.

Data presented in Tables II and III indicate for each clone the most stimulatory peptides of either hHSP60 or CpHSP60. However, a num-ber of clones proliferated at lower degree also in response to one or both of the adjacent peptides, suggesting that their specific epitope was present in more than one single peptide (Fig. 2).

To know the MHC restriction elements required for Ag recog-nition, hHSP60-specific T cell clones were stimulated by hHSP60 or CpHSP60 in the presence of irradiated autologous APCs treated with anti-HLA-DR or anti-HLA-DQ mAbs. Anti-HLA-DR sulted consistently in virtual abrogation of the proliferative re-sponse by T cell clones to either hHSP60 or CpHSP60, whereas anti-HLA-DQ was unable to affect self- or cross-reactive HSP60-induced T cell clone proliferation (data not shown). Because anti-HLA-DR Abs may hamper T cell responses, even if the specific

response under study is not MHC class II restricted, T cell clones were stimulated with the relevant hHSP60 peptide in the presence of either autologous-irradiated PHA-induced T cell blasts or allo-geneic-irradiated T cell blasts from donors matched for one single

DRB1 allele. As shown in Table V, for most of plaque-derived

clones, the proliferative response to the specific peptide was re-stricted by one of the DRB1 alleles, whereas for a few clones, both

DRB1 alleles allowed peptide presentation.

Functional profile of autoreactive and cross-reactive hHSP60-specific T cell clones

All plaque-derived hHSP60-specific clones were assessed for their cytokine profile on Ag stimulation. In the series of hHSP60-spe-cific clones not cross-reactive to CpHSP60, 22 (84.6%) secreted IFN-␥ and TNF-␣ but not IL-4 (Th1 profile), whereas in 4 clones, stimulation with hHSP60 resulted in the production of IL-4 as well (Th0 profile).

Likewise, in the series of 18 hHSP60/CpHSP60 cross-reactive clones, stimulation with either hHSP60 or CpHSP60 disclosed a Th1 profile in 15 clones (83.3%) and a Th0 profile in the other 3. The cytolytic potential of hHSP60-specific autoreactive or cross-reactive T cell clones was assessed by using Ag-pulsed51

Cr-labeled autologous EBV-B cells as targets. At an E:T ratio of 10:1, 36 of the 37 (97%) Th1 and 5 of 7 (71%) Th0 clones lysed hHSP60-presenting autologous EBV-B cells (range of specific

51_{Cr release, 18 – 63%), whereas autologous EBV-B cells pulsed}

with hHSP70 (control) Ag and cocultured with the same clones Table IV. Cross-reactivity to MbHSP65 and GroEL or to their peptides of CD4 T cell clones reactive to both hHSP60 and

CpHSP60 obtained from the atherosclerotic plaques of anti-C. pneumoniae seropositive patientsa

T Cell Clone Ag or Peptide (position and sequence)

Proliferative Response (MI)a

10␮g/ml 1␮g/ml 0.1␮g/ml 5.05 hHSP60 43 19 ⬍ 2 hHSP60 76–90 DGVTVAKSIDLKDKY 98 56 23 CpHSP60 68 29 ⬍ 2 CpHSP60 51–65 KDGVTVAKEIELEDK 92 61 31 MbHSP65 8 ⬍ 2 ⬍ 2 MbHSP65 51–65 DGVSIAKEIELEDPY 21 ⬍ 2 ⬍ 2 GroEL 13 ⬍ 2 ⬍ 2 GroEL 51–65 KDGVSVAKEIELEDK 28 ⬍ 2 ⬍ 2 5.16 hHSP60 93 34 6 hHSP60 166–180 EEIAQVATISANGDK 138 71 39 CpHSP60 68 29 ⬍ 2 CpHSP60 141–155 HKEIAQVATISANND 101 66 23 MbHSP65 6 ⬍ 2 ⬍ 2 MbHSP65 141–155 QIAATAAISAGDQSI 11 ⬍ 2 ⬍ 2 GroEL 17 ⬍ 2 ⬍ 2 GroEL 141–155 SEEVAQVGTISANGD 31 2 ⬍ 2 8.22 hHSP60 40 19 ⬍ 2 hHSP60 51–65 VAVTMGPKGRTVIIE 69 37 25 CpHSP60 31 14 ⬍ 2 CpHSP60 26–40 AVKVTLGPKGRHVVI 62 29 11 MbHSP65 16 ⬍ 2 ⬍ 2 MbHSP65 26–40 KVTLGPKGRNVVLEK 22 ⬍ 2 ⬍ 2 GroEL 5 ⬍ 2 ⬍ 2 GroEL 26–40 AVKVTLGPKGRNVII 31 2 ⬍ 2 8.26 hHSP60 151 78 22 hHSP60 171–185 VATISANGDKEIGNI 129 84 37 CpHSP60 113 65 30 CpHSP60 146–160 QVATISANNDSEIGN 165 91 36 MbHSP65 ⬍ 2 ⬍ 2 ⬍ 2 MbHSP65 146–160 TAA ISAGDQSIGDLI 9 ⬍ 2 ⬍ 2 GroEL 8 ⬍ 2 ⬍ 2 GroEL 146–160 QVATISANGDKQVGL 22 ⬍ 2 ⬍ 2 a

Culture conditions are reported in the legend to Table II. Underlined letters indicate the amino acids shared between hHSP60 and the other HSP60.

Bold letters indicate amino acid changes apparently crucial for peptide recognition by T cell clones 5.05 and 8.22.

by guest on February 6, 2017

http://www.jimmunol.org/

were not lysed. The relative potency of Ag-induced cytotoxic ac-tivity of HSP60-specific T cell clones against autologous EBV-B cells pulsed with hHSP60 or CpHSP60 was assessed by compar-ison of levels of the specific 51_{Cr release at different E:T ratios}

(Fig. 3). Because activated effector T cells can also kill their targets by inducing apoptosis through Fas-Fas ligand interaction (17, 20), we evaluated the ability of activated hHSP60-specific clones to induce51_{Cr release by Fas}⫹_{Jurkat cells undergoing apoptosis. On}

mitogen activation, 32 of 37 Th1 (86%) and 4 of 7 (57%) Th0 clones were able to induce apoptosis in target cells (range of spe-cific51_{Cr release, 21–59%).}

hHSP60-activated plaque-infiltrating T cells help monocyte TF production

Because plaque rupture and thrombosis are notable complications of atherosclerosis, we asked whether stimulation with hHSP60 or CpHSP60 might enable plaque-infiltrating autoreactive or cross-reactive T cells to express helper function for TF production by monocytes. All clones were cocultured with autologous monocytes in the absence or presence of medium alone, hHSP70, hHSP60, or CpHSP60, and TF protein was measured. In the presence of me-dium alone or hHSP70, none of the 44 plaque-derived clones ex-pressed helper function for monocyte TF production, ruling out the possibility that monocytes expressed hHSP60 suitable for T cell activation (Fig. 4). In contrast, apart from 2 Th0 clones (one au-toreactive and one cross-reactive), in 42 (95%) hHSP60-specific clones, stimulation with hHSP60 resulted in the expression of sub-stantial help for TF production by monocytes. Likewise, stimula-tion with CpHSP60 enabled 17 of 18 (94%) cross-reactive T cell clones to induce monocyte TF production, whereas it failed to elicit any helper function for TF production in the 25 autoreractive clones that proliferated in response to hHSP60 but not to CpHSP60 (Fig. 4).

Discussion

A number of Ags are suspected to play a role in atherosclerosis-associated immune reactions. Among other candidates, the list in-cludes modified LDL, such as oxidized low-density lipoprotein (21), viral, and bacterial components, and a family of phylogeneti-cally highly conserved proteins known as HSPs, which are ex-pressed in prokaryotic and eukaryotic cells under physiological conditions and in response to various forms of stress. Under both normal and stressed conditions, HSPs, as molecular chaperones, facilitate the folding of nascent proteins, resolubilize protein ag-gregates, assist in refolding denatured proteins (22), and help in molecular transport across the intracellular membranes (23). Cells of the arterial wall are stimulated to produce high levels of HSPs in response to a number of factors including infections, fever, me-chanical or oxidant stress, or exposure to cytokines, heavy metals, alcohol, or inhibitors of energy metabolism (24). Studies in exper-imental models indicate that 60- to 65-kDa HSP may play a proatherogenic role. Atherosclerosis-like lesions were induced in normocholesterolaemic rabbits by immunization with mycobacte-ria or recombinant mycobactemycobacte-rial HSP65 (25), and T cells isolated from these lesions were found to respond specifically to HSP65 in vitro (26, 27).

In this study, we demonstrate that atherosclerotic patients har-bored in their carotid plaques in vivo-activated CD4⫹T cells that reacted specifically to self HSP60. In addition, all four patients with positive serology and PCR detection of C. pneumoniae DNA had in their carotid plaques at least two populations of hHSP60-specific T cells: one reactive only to self hHSP60, and the other reactive to both the self and the C. pneumoniae analog HSP60. Blocking experiments with anti-DR and anti-DQ Abs and cocul-ture of T cell clones with appropriate allogeneic APC showed that

FIGURE 1. Dose-response effect of graded Ag or peptide concentra-tions on the proliferative response to hHSP60 and CpHSP60 of plaque-derived T cell clones. T cell blasts from each clone were cocultured with autologous-irradiated APC in the presence of graded concentrations of hHSP60 (F), CpHSP60 (E), or of the appropriate hHSP60 peptides (f), or CpHSP60 peptides (䡺). Results represent mean values of MIs measured in quadruplicate cultures of 8 representative of 26 clones reactive only to hHSP60 (upper and middle panels) and in 4 representative of 18 clones reactive to both hHSP60 and CpHSP60 (lower panels).

FIGURE 2. Proliferative response of hHSP60-specific T-clones to se-ries of partially overlapping peptides of hHSP60, including the specific epitopes. T cell blasts from clones derived from the plaques of four patients sharing the DRB1*07 allele were cocultured with autologous-irradiated APC in the presence of series of 3 partially overlapping 15-mer peptides, including the most stimulatory one, and a couple of adjacent peptides. Results represent mean values of MIs measured in quadruplicate cultures of eight representative T cell clones.

by guest on February 6, 2017

http://www.jimmunol.org/

DR represents the MHC restriction element in the T cell response to either hHSP60 or CpHSP60. It is of note that, despite good viability and IL-2-induced growth, PBMC-derived T cell clones from all patients consistently failed to proliferate in response to hHSP60, CpHSP60, or hHSP70. In the absence of obvious expla-nations due to technical pitfalls, the reason for the inability to

detect hHSP-specific T cells in the peripheral blood remains un-clear. One possibility is that hHSP60-specific T cells are present in the peripheral blood in resting state, like most of circulating T cells, and are not suitable for in vitro activation and expansion induced by IL-2. Another possibility is that very low numbers, if any, of in vivo-activated hHSP60-specific T cells are present in the peripheral blood, whereas they concentrate into the lesions of ar-terial walls, where they find their specific Ag(s) and participate in the pathology of atherosclerosis by expressing their effector mech-anisms. That such activated T cells infiltrating the atherosclerotic

FIGURE 3. Dose-response effect of graded E:T ratios on the cytotoxic activity of plaque-derived T cell clones specific for hHSP60 against Ag-pulsed autologous APC. To assess their perforin-mediated cytotoxicity, T cell clones reactive to hHSP60 (A) or to both hHSP60 and CpHSP60 (B) were cocultured at different E:T ratios with 51_{Cr-labeled autologous}

EBV-B cells pulsed with hHSP70 (10␮g/ml) (䡺) or hHSP60 (10 ␮g/ml) (f) or CpHSP60 (10␮g/ml) (Œ), and51_{Cr release was measured as index}

of Ag-induced specific target cell lysis. Results represent mean values of

51_{Cr release measured in triplicate cultures of eight representative T cell}

clones.

FIGURE 4. Monocyte TF production induced by plaque-derived T cell clones stimulated with hHSP60 or CpHSP60. T cell blasts from each clone were cocultured with autologous monocytes in the presence of hHSP70 (control Ag) or the specific Ag (hHSP60 or CpHSP60), and TF production was assessed by a specific ELISA.

Table V. MHC class II restriction of proliferative response of plaque-derived T cell clones reactive to

selected hHSP60 peptidesa

Code and MHC Class II Haplotype of Clones MHC Class II Haplotype of Irradiated APC MI in Responsea to hHSP60 Peptide (position) 2.08 DRB1ⴱ03–07 Autologous 108 (31– 45) DRB1ⴱ03 111 DRB1ⴱ07–11 3 2.40 DRB1ⴱ03–07 Autologous 97 (136 –150) DRB1ⴱ03 88 DRB1ⴱ07–11 63 2.53 DRB1ⴱ03–07 Autologous 66 (321–335) DRB1ⴱ03 7 DRB1ⴱ07–11 52 3.40 DRB1ⴱ01–16 Autologous 71 (91–105) DRB1ⴱ01–04 4 DRB1ⴱ11–16 65 3.43 DRB1ⴱ01–16 Autologous 36 (206 –220) DRB1ⴱ01–04 29 DRB1ⴱ11–16 2 4.31 DRB1ⴱ04–11 Autologous 54 (446 – 480) DRB1ⴱ01–04 ⬍ 2 DRB1ⴱ11–16 49 4.25 DRB1ⴱ04–11 Autologous 112 (491–505) DRB1ⴱ01–04 89 DRB1ⴱ11–16 3 7.18 DRB1ⴱ03–07 Autologous 68 (46 – 60) DRB1ⴱ03 79 DRB1ⴱ07–11 ⬍ 2 Autologous 59 (21–35 of CpHSP60) DRB1ⴱ03 41 DRB1ⴱ07–11 17 a T cell blasts (105

/well) from each clone were cocultured with irradiated autologous or allogeneic activated T cell blasts as APC (105

/well) in the presence of medium alone or the specific hHSP60 or CpHSP60 peptide (10␮g/ml), and proliferative responses (MI) were measured after 3 days. Results are reported as mean values obtained in quadruplicate cultures, SD values being⬍ 16% of means.

by guest on February 6, 2017

http://www.jimmunol.org/

plaques actually participate in the disease pathogenesis is sup-ported by the observation that the formation of arteriosclerotic le-sions by immunization of rabbits with HSP65-containing material could be abolished by immunosuppression and T cell depletion with an anti-CD3 Ab plus prednisolone (27).

The presence in the plaques of Cp-positive patients of T cells reactive to CpHSP60 is in agreement with the observations by other laboratories (28, 29) and with our earlier demonstration in a similar series of patients of T cells reactive to C. pneumoniae Ags, such as sonicated elementary bodies, the 10 kDa HSP, the outer membrane protein 2, and the CpHSP60 (9). In the present study, however, the presence of plaque-derived T cell clones specific for

C. pneumoniae Ags different from CpHSP60 was not investigated.

T cell recognition of either hHSP60 or CpHSP60 resulted in both proliferation and expression of functional properties by T cell clones, i.e., a predominant Th1 profile. In addition, on appropriate stimulation, the great majority of plaque-derived HSP60-specific clones induced both perforin-mediated cytolysis and Fas-Fas li-gand-mediated apoptosis in target cells. Based on these findings, it is tempting to hypothesize that in the inflammatory setting of the atherosclerotic plaque in which HSP60-specific autoreactive or cross-reactive Th1 cells are activated, endothelial cells may ac-quire APC function for HSP60 and, together with professional APCs (30), can become targets of the cytotoxic and proapoptotic activity of HSP60-specific Th1 cells. The outcome of this process would be the expansion of the plaque and the formation of the necrotic cores characteristic of complicated and unstable athero-sclerotic lesions. A linkage has been suggested between the degree of macrophage apoptosis and plaque rupture, to which apoptotic death of smooth muscle cells may also contribute (31–33). More-over, it is reasonable to suspect that HSP60-activated Th1 cells and their cytokines can play a role in driving the up-regulation of TF production by monocytes within atherosclerotic plaques, thus con-tributing to the thrombogenicity of lesions (34). Indeed, the Th1 polarization of T cell responses and the poor production of Th2 cytokines occurring within the plaque may represent local risk fac-tors of thrombosis (19), which associate with platelet adhesion to dysfunctional endothelium.

Our findings support the hypothesis that a crucial component of atherosclerosis is represented by Th1 cell-mediated immune re-sponses to self and/or foreign Ags. More than 95% sequence ho-mology exists between HSP60s from various bacteria, and even between bacterial and hHSP60 a 50 –55% sequence homology ex-ists, and in highly conserved regions it reaches⬎70% (35). The analysis of the submolecular specificity of T cell clones reactive only to hHSP60 and of clones reactive to both hHSP60 and CpHSP60 showed that the former recognized their epitope in por-tions of relatively poor or no homology between the two proteins, whereas the latter found their specific epitope in regions of high-sequence homology. Therefore, a number of hHSP60 T cell epitopes are “private,” such as the 1–15 and 6 –20 N-terminal or the 506 –520 C-terminal sequences recognized by different clones of different Cp-negative patients and by a few clones of Cp-posi-tive patients, whereas other hHSP60 epitopes are similar to, and cross-reactive with, T cell epitopes of CpHSP60. T cell clones specific for private or cross-reactive epitopes of hHSP60 do, how-ever, express similar predominant Th1 profile and may contribute equally to inflammation in the setting of atherosclerosis.

HSP60 are released in soluble form from the surface of stressed or damaged cells and can be found in the supernatant of cell cul-tures in vitro or in the serum in vivo (36, 37). Whether soluble hHSP60 released in the inflammatory setting of the plaque may undergo biochemical changes, becoming the target of bona fide autoimmunity (3), remains to be investigated. Likewise, it remains

to be established whether autoimmunity to hHSP60 results from a breakdown of tolerance associated with chronic inflammation and aging, and what is the role of the molecular mimicry between hHSP60 and the HSP60 of pathogens, such as C. pneumoniae, detected in the atherosclerotic lesions. The mycobacterial homo-logue MbHSP65 and the E. coli homohomo-logue GroEL (38) might be other candidates for cross-reactive recognition by hHSP60-reac-tive T cell clones in atherosclerotic plaques. In this study, however, none of the 26 T cell clones reactive to hHSP60, but not to CpHSP60, also reacted to MbHSP65 or to GroEL. Only 11 and 9 of the 18 T cell clones reactive to both hHSP60 and CpHSP60 showed poor or negligible response to MbHSP65 or GroEL, re-spectively, on the basis of dose-response curves. Neither MbHSP65 nor GroEL peptides that partially overlapped specific hHSP60/CpHSP60 epitopes were able to induce T cell clone pro-liferation at intermediate or low peptide concentrations, arguing against the hypothesis that MbHSP65 or GroEL might represent major targets of plaque-infiltrating T cells in our patients.

Data obtained in this study support the hypothesis that two ma-jor mechanisms, partially overlapping and not mutually exclusive, may be responsible for the T cell-mediated immunopathology of atherosclerosis. The first one would imply that arterial endothelial cells, undergoing the effects of classical stress factors associated with atherosclerosis and conditioned by cytokines produced by plaque-infiltrating Th1 cells, express self 60-kDa HSP. Such an autoantigen would be presented by professional APC and endo-thelial cells, becoming a target of autoreactive T cells specific for private epitopes of hHSP60. T cell-mediated cytotoxic and apo-ptotic killing of stressed endothelial cells expressing self HSP60 may activate a vicious circle of self maintenance of such Th1-mediated autoimmune mechanism of endothelial damage. The sec-ond mechanism, active in patients who failed to clear C.

pneu-moniae, would be mediated by plaque-infiltrating Th1 cells

specific for C. pneumoniae Ags, among which CpHSP60-specific T cells that cross-recognize shared epitopes of the hHSP60 via a mechanism of molecular mimicry. The availability of self HSP60 expressed by the vascular endothelium would contribute to a sec-ond branch of the vicious circle of self maintenance of the immune response that would be mediated by CpHSP60-specific Th1 cells that cross-react to self HSP60. These possibilities are consistent with the results obtained in several experimental animal models. In these models, a central role for T cells specific for HSP60s has indeed been established. Immunization with HSP65 of LDL re-ceptor-deficient (LDL-R⫺/⫺) mice induced specific T cell reactiv-ity against HSP65 as well as mammalian HSP60, and transfer into nonimmunized mice of lymphocytes or purified IgG of immunized animals enhanced the size of vascular lesions (39). In contrast, nasal or oral immunization with HSP65 of hypercholesterolemic ApoE⫺/⫺and LDL-R⫺/⫺mice resulted in reduced T cell reactivity to HSP and attenuated atherosclerosis (40, 41). In contrast, arthri-togenic and arthritis-preventing HSP60 epitopes have been iden-tified in the model of adjuvant arthritis in rats (42, 43) and human rheumatoid arthritis (44). The identification in present study of a number of atherosclerosis-associated T cell epitopes of HSP60 may be of importance for designing strategies on preventive or therapeutic approaches aimed to inhibit the immune and autoim-mune pathogenic mechanisms of atherosclerosis.

Disclosures

The authors have no financial conflict of interest.

References

1. Saikku, P., M. Leinonen, K. Mattila, M. R. Ekman, M. S. Nieminen, P. H. Makela, J. K. Huttunen, and V. Valtonen. 1988. Serological evidence of an

by guest on February 6, 2017

http://www.jimmunol.org/

association of a novel Chlamydia, TWAR, with chronic coronary heart disease and acute myocardial infarction. Lancet 2: 983–986.

2. Shi, W., N. J. Wang, D. M. Shih, V. Z. Sun, X. Wang, and A. J. Lusis. 2000. Determinants of atherosclerosis susceptibility in the C3H and C57BL/6 mouse model: evidence for involvement of endothelial cells but not blood cells or cho-lesterol metabolism. Circ. Res. 86: 1078 –1084.

3. Wick, G., M. Knoflach, and Q. Xu. 2004. Autoimmune and inflammatory mech-anisms in atherosclerosis. Annu. Rev. Immunol. 22: 361– 403.

4. Ross, R. 1999. Atherosclerosis: an inflammatory disease. N. Engl. J. Med. 340: 115–126.

5. Epstein, S. E., Y. F. Zhou, and J. Zhu. 1999. Infection and atherosclerosis: emerg-ing mechanistic paradigms. Circulation 100: 20 –28.

6. Jonasson, L., J. Holm, O. Skalli, G. Bondjers, and G. K. Hansson. 1986. Regional accumulations of T cells, macrophages, and smooth muscle cells in the human atherosclerotic plaque. Arteriosclerosis 6: 131–138.

7. Hansson, G. K., L. Jonasson, P. S. Seifert, and S. Stemme. 1989. Immune mech-anisms in atherosclerosis. Arteriosclerosis 9: 567–578.

8. Kleindienst, R., Q. Xu, J. Willeit, F. R. Waldenberger, S. Weimann, and G. Wick. 1993. Immunology of atherosclerosis: demonstration of heat shock protein 60 expression and T lymphocytes bearing␣/␤ or ␥/␦ receptor in human atheroscle-rotic lesions. Am. J. Pathol. 142: 1927–1937.

9. Benagiano, M., A. Azzurri, A. Ciervo, A. Amedei, C. Tamburini, M. Ferrari, J. L. Telford, C. T. Baldari, S. Romagnani, A. Cassone, et al. 2003. T helper type-1 lymphocyte-driven inflammation in human atherosclerotic lesions. Proc.

Natl. Acad. Sci. USA 100: 6658 – 6663.

10. Khan, M. A., and C. W. Potter. 1996. The nPCR detection of Chlamydia

pneu-moniae and Chlamydia trachomatis in children hospitalized for bronchiolitis. J. Infect. 33: 173–175.

11. Del Prete, G., M. De Carli, M. M. D’Elios, P. Maestrelli, M. Ricci, L. Fabbri, and S. Romagnani. 1993. Allergen exposure induces the activation of allergen-spe-cific Th2 cells in the airway mucosa of patients with allergic respiratory disor-ders. Eur. J. Immunol. 23: 1445–1449.

12. D’Elios, M. M., M. Manghetti, F. Almerigogna, A. Amedei, F. Costa, D. Burroni, C. T. Baldari, S. Romagnani, J. L. Telford, and G. Del Prete. 1997. Different cytokine profile and antigen-specificity repertoire in Helicobacter pylori-specific T cell clones from the antrum of chronic gastritis patients with or without peptic ulcer. Eur. J. Immunol. 27: 1751–1755.

13. Amedei, A., M. P. Bergman, B. J. Appelmelk, A. Azzurri, M. Benagiano, C. Tamburini, R. van der Zee, J. L. Telford, C. M. Vandenbroucke-Grauls, M. M. D’Elios, and G. Del Prete. 2003. Molecular mimicry between Helicobacter pylori antigens and H⫹,K⫹-ATPase in human gastric autoimmunity. J. Exp. Med. 198: 1147–1156.

14. Ciervo, A., P. Visca, A. Petrucca, L. M. Biasucci, A. Maseri, and A. Cassone. 2002. Antibodies to 60-kilodalton heat shock protein and outer membrane protein 2 of Chlamydia pneumoniae in patients with coronary heart disease. Clin. Diagn.

Lab. Immunol. 9: 66 –74.

15. Van der Zee, R., S. M. Anderton, C. A. F. Buskens, E. Alonso de Velasco, and W. Van Eden. Heat shock protein T cell epitopes as immunogenic carriers in subunit vaccines. In Peptides 1994, Proceedings of the Twenty-Third Peptide

Symposium. H. L. S. Maya, ed. ESCOM, Leiden, p. 841.

16. Del Prete, G., M. De Carli, M. Ricci, and S. Romagnani. 1991. Helper activity for immunoglobulin synthesis of T helper type 1 (Th1) and Th2 human T cell clones: the help of Th1 clones is limited by their cytolytic capacity. J. Exp. Med. 174: 809 – 813.

17. Vergelli, M., B. Hemmer, P. A. Muraro, L. Tranquilli, W. E. Biddison, A. Sarin, H. F. McFarland, and R. Martin. 1997. Human autoreactive CD4⫹T cell clones use perforin- or Fas/Fas ligand-mediated pathways for target cell lysis. J.

Immu-nol. 158: 2756 –2761.

18. D’Elios, M. M., M. P. Bergman, A. Azzurri, A. Amedei, M. Benagiano, J. J. De Pont, F. Cianchi, C. M. Vandenbroucke-Grauls, S. Romagnani, B. J. Appelmelk, and G. Del Prete. 2001. H⫹,K⫹-atpase (proton pump) is the target autoantigen of Th1-type cytotoxic T cells in autoimmune gastritis.

Gas-troenterology 120: 377–386.

19. Del Prete, G., M. De Carli, R. M. Lammel, M. M. D’Elios, K. C. Daniel, B. Giusti, R. Abbate, and S. Romagnani. 1995. Th1 and Th2 T-helper cells exert opposite regulatory effects on procoagulant activity and tissue factor production by human monocytes. Blood 86: 250 –257.

20. Kagi, D., F. Vignaux, B. Ledermann, K. Burki, V. Depraetere, S. Nagata, H. Hengartner, and P. Golstein. 1994. Fas and perforin pathways as major mech-anisms of T cell-mediated cytotoxicity. Science 265: 528 –530.

21. Stemme, S., B. Faber, J. Holm, O. Wiklund, J. L. Witztum, and G. K. Hansson. 1995. T lymphocytes from human atherosclerotic plaques recognize oxidized low density lipoprotein. Proc. Natl. Acad. Sci. USA 92: 3893–3897.

22. Schlesinger, M. J. 1990. Heat shock proteins. J. Biol. Chem. 265: 12111–12114. 23. Morimoto, R. I. 1993. Cells in stress: transcriptional activation of heat shock

genes. Science 259: 1409 –1410.

24. Lamb, D. J., W. El-Sankary, and G. A. A. Ferns. 2003. Molecular mimicry in atherosclerosis: a role for heat shock proteins in immunisation. Atherosclerosis 167: 177–185.

25. Xu, Q., H. Dietrich, H. J. Steiner, A. M. Gown, B. Schoel, G. Mikuz, S. H. Kaufmann, and G. Wick. 1992. Induction of atherosclerosis in normocho-lesterolaemic rabbits by immunisation with heat shock protein 65. Arterioscler.

Thromb. 12: 789 –799.

26. Xu, Q., R. Kleindienst, W. Waitz, H. Diertrich, and G. Wick. 1993. Increased expression of heat shock protein 65 coincides with a population of infiltrating T lymphocytes in atherosclerotic lesions of rabbits specifically responding to heat shock protein 65. J. Clin. Invest. 91: 2693–2702.

27. Metzler, B., M. Mayr, H. Dietrich, M. Singh, E. Wiebe, Q. Xu, and G. Wick. 1999. Inhibition of arteriosclerosis by T-cell depletion in normocholesterolaemic rabbits immunised with heat shock protein 65. Arterioscler. Thromb. Vasc. Biol. 19: 1905–1911.

28. Curry, A. J., I. Portig, J. C. Goodall, P. J. Kirkpatrick, and J. S. Gaston. 2000. T lymphocyte lines isolated from atheromatous plaque contain cells capable of responding to Chlamydia antigens. Clin. Exp. Immunol. 121: 261–269. 29. Mosorin, M., H. M. Surcel, A. Laurila, M. Lehtinen, R. Karttunen, J. Juvonen,

J. Paavonen, R. P. Morrison, P. Saikku, and T. Juvonen. 2000. Detection of

Chlamydia pneumoniae-reactive T lymphocytes in human atherosclerotic plaques

of carotid artery. Arterioscler. Thromb. Vasc. Biol. 20: 1061–1067.

30. Millonig, G., C. Schwentner, P. Mueller, C. Mayerl, and G. Wick. 2001. The vascular-associated lymphoid tissue: a new site of local immunity. Curr. Opin.

Lipidol. 12: 547–553.

31. Kolodgie, F. D., J. Narula, A. P. Burke, N. Haider, A. Farb, Y. Hui-Liang, J. Smialek, and R. Virmani. 2000. Localization of apoptotic macrophages at the site of plaque rupture in sudden coronary death. Am. J. Pathol. 157: 1259 –1268. 32. Geng, Y. J., L. E. Henderson, E. B. Levesque, M. Muszynski, and P. Libby. 1997. Fas is expressed in human atherosclerotic intima and promotes apoptosis of cy-tokine-primed human vascular smooth muscle cells. Arterioscler. Thromb. Vasc.

Biol. 17: 2200 –2208.

33. Mallat, Z., and A. Tedgui. 2001. Current perspective on the role of apoptosis in atherothrombotic disease. Circ. Res. 88: 998 –1003.

34. Toschi, V., R. Gallo, M. Lettino, J. T. Fallon, S. D. Gertz, A. Fernandez-Ortiz, J. H. Chesebro, L. Badimon, Y. Nemerson, V. Fuster, and J. J. Badimon. 1997. Tissue factor modulates the thrombogenicity of human atherosclerotic plaques.

Circulation 95: 594 –599.

35. Young, R. A., and T. J. Elliott. 1989. Stress proteins, infection, and immune surveillance. Cell 59: 5– 8.

36. Liao, D. F., Z. G. Jin, A. S. Baas, G. Daum, S. P. Gygi, R. Aebersold, and B. C. Berk. 2000. Purification and identification of secreted oxidative stress-induced factors from vascular smooth muscle cells. J. Biol. Chem. 275: 189 –196. 37. Xu, Q., G. Schett, H. Perschinka, M. Mayr, G. Egger, F. Oberhollenzer, J. Willeit, S. Kiechl, and G. Wick. 2000. Serum soluble heat shock protein 60 is elevated in subjects with atherosclerosis in a general population. Circulation 102: 14 –20. 38. Hartl, F. U., and M. Hayer-Hartl. 2002. Molecular chaperones in the cytosol:

from nascent chain to folded protein. Science 295: 1852–1858.

39. George, J., A. Afek, B. Gilburd, Y. Shoenfeld, and D. Harats. 2001. Cellular and humoral immune responses to heat shock protein 65 are both involved in pro-moting fatty-streak formation in LDL-receptor deficient mice. J. Am. Coll.

Car-diol. 38: 900 –905.

40. Harats, D., N. Yacov, B. Gilburd, Y. Shoenfeld, and J. George. 2002. Oral tol-erance with heat shock protein 65 attenuates Mycobacterium tuberculosis-in-duced and high-fat-diet-driven atherosclerotic lesions. J. Am. Coll. Cardiol. 40: 1333–1338.

41. Maron, R., G. Sukhova, A. M. Faria, E. Hoffmann, F. Mach, P. Libby, and H. L. Weiner. 2002. Mucosal administration of heat shock protein-65 decreases atherosclerosis and inflammation in aortic arch of low-density lipoprotein recep-tor-deficient mice. Circulation 106: 1708 –1715.

42. Barker, R. N., A. J. Easterfield, R. F. Allen, A. D. Wells, C. J. Elson, and S. J. Thompson. 1996. B- and T-cell autoantigens in pristane-induced arthritis.

Immunology 89: 189 –194.

43. Anderton, S. M., R. Van der Zee, B. Prakken, A. Noordzij, and W. van Eden. 1995. Activation of T cells recognizing self 60-kD heat shock protein can protect against experimental arthritis. J. Exp. Med. 181: 943–952.

44. Gaston, J. S. 1998. Role of T-cells in the development of arthritis. Clin. Sci. 95: 19 –31.

by guest on February 6, 2017

http://www.jimmunol.org/